allan-fraser avatar image
allan-fraser asked

SmartSolar goes to Float before batteries are fully charged.

While two of my three SmartSolar 100/30s (A & B) stay in Bulk all day (at this time of year), the third (C) goes into Absorption just before noon and then into Float roughly 2.5 hours later. Consequently my batteries never get beyond 88% charged. (If I disable the inputs to A and B, C remains in Bulk all day.)

SmartSolars A and B are each connected to 2 * 135w panels in series; C is connected to 2 * 170w panels also in series. The battery bank is 4 * Lifeline 125Ah AGMs in parallel.

All three SmartSolars have the latest firmware (v1.61) and the same custom profile:

Absorption voltage: 14.3 (per Lifeline’s recommendation)

Float voltage: 13.3 (per Lifeline’s recommendation)

Adaptive absorption time: ON

Maximum absorption time: 6hrs

Re-bulk voltrage offset: -0.4v

Tail current: 0.1A

Expert mode: OFF


  1. Can anyone explain why C is going to Absorption while A and B remain in Bulk and what can I do to stop C going to Float until the batteries are fully charged?
  2. Would setting up the SmartSolars in a VE.Smart Network so that their charging is synchronised help? If so, can I do this without a Smart Battery Sense or BMV in the network?
smart solar charging behaviourVE.Smart Network
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
Rob Fijn avatar image
Rob Fijn answered ·

Could be a bad or lose contact in the Battery cable connection. Between the MPPT and your bus-bar.
You can check this via the ConnectApp, compare the battery voltage. When the charge current is high and there is a bad /lose connection, the battery voltage reading at the MPPT goes up... so the MPPT 'thinks' then battery is full and goes in to absorption..

Let us know what you find.


2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

allan-fraser avatar image allan-fraser commented ·

I’m running the battery bank down over night and will disable A & B then monitor the voltages as the charge current ramps up. I’ll let you know.

0 Likes 0 ·
allan-fraser avatar image allan-fraser allan-fraser commented ·

At around midday, with just C in play and pumping out 16-18A at ~13.6v (still bulk mode), the voltage drop between the MPPT controller and the busbar was 0.08-0.11v. The cable length is 700mm / just over 2ft. All of the connections were tight. I don’t know if 0.1v is significant but checking this out reminded me that the wiring I put in (6mm2) was at the lower limit of what was required and that upping this to 10mm2 is still on my to do list. Thanks!

0 Likes 0 ·
snoobler avatar image
snoobler answered ·

What is reporting 88% charge?

Expert mode "OFF" doesn't turn off these settings, it just doesn't display them. You need to confirm that these are all the same as well.

0.1A tail current is likely far too low for the batteries. Most AGM are fully charged at absorption voltage when the current has dropped to 0.01-0.02C or 5-10A in your case. If you're hitting 14.3V, holding it, and the total charge current has dropped to 5-10A, you're getting fully charged.

1) with no charger coordination or any shared voltage and/or current sensing device, the voltages sensed by the chargers dictates their behavior. Since "C" has the highest power, it is likely "seeing" 14.3V first, so it hits absorption first. As A & B catch up, it can no longer deliver it's current, hits tail current and drops to float.

2) Yes, but you need a SBS, smartshunt or BMV-712 to establish the network.

With a VE.Smart network, an SBS would improve function somewhat by providing shared voltage and temperature data. A smartshunt with temp sensor would improve function substantially due to the shunt feeding both voltage and current data to the MPPTs.

Note that since you're using AGM batteries, you should have a temperature sensor feeding the MPPTs suitable temperature data for temperature compensated charging.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

allan-fraser avatar image allan-fraser commented ·

The 88% is coming from a shunt-based Philippi BMS that I’ve been using for over three years now. During the summer months the batteries regularly get to 100% although it’s possible I hadn’t noticed C going to Absorption and Float early and that it has been A & B getting the batteries over the line.

Currently, around the time C goes to Absorption, C is delivering 10-12A while A & B are delivering 7-8A each. And when C goes to Float, it then delivers 0A while A & B are still delivering 7-8A each.

At the moment I don’t have the additional kit needed to set up a VE.Smart Network. However, in order to get a more integrated and comprehensive view of what’s happening, I am planning to replace the Philippi BMS later this year and install VE SmartShunts (and now temp sensors) and a Cerbo GX & display so I will do the network then.

Meanwhile, would there be any benefit in changing the charge profile on SmartSolar C from adaptive to fixed with a period of, say, 6 hours to stop C going to Float early?

0 Likes 0 ·
snoobler avatar image snoobler allan-fraser commented ·
Probably not. Drop to float is triggered by tail current and the adaptive absorption algorithm.

As was mentioned by @Rob Fijn , triple checking all your wires/connections, etc. for proper size, torque and quality is always a good idea.

For experimentation, I would set all three to 6 hours fixed absorption and 5A tail current.

I would also review the criteria by which your "BMS" is computing state of charge. I suspect it may not be configured to properly sync at 14.3V

14.3V is on the lower end of typical absorption voltages and may simply require a much longer absorption period. Given that you're maintaining 14.3V, and the sum of current is about 14A, you're likely getting really darn close to 100%. This voltage may ultimately be the limiting factor as it's going to increase your charge time.

The chargers' voltage readings are influenced by the current they are sending to the battery, their wire gauge and lengths. A SBS is all of about $50 and would certainly improve things by:

1) Give all chargers an accurate voltage measurement

2) Allowing for proper charging with temperature compensation.

3) Allow the chargers to coordinate in a VE.Smart network.

0 Likes 0 ·
allan-fraser avatar image allan-fraser snoobler commented ·

Philippi doesn’t provide much info on how its software determines the SoC but the system has been working fine for almost four years, regularly getting to 100% no matter what the charging source: solar, alternator or shore power. I’m on shore power this evening so will see what SoC the Philippi reports tomorrow morning.

The SBS looks like a good interim solution but as I’m in Tunisia with no prospect of finding one here or getting one shipped in, it will need to wait until I’m back on mainland Europe. Meanwhile I’ll try out the 6 hour fixed absorption option and report back fyi.

0 Likes 0 ·
snoobler avatar image snoobler allan-fraser commented ·

I check out a couple of the Philippi options, and they generally look pretty good; however, they must be programmed specifically for your charge cycle, so unless you're 100% confident it's programmed correctly, 100% SoC may not mean 100%. All battery monitors' accuracy drifts, and the need to be synchronized to 100% regularly to maintain accuracy.

Generally speaking, your battery is fully charged once 14.3V is reached AND it is held there long enough for the current to drop to 5-10A. The Philippi should be programmed similarly.

0 Likes 0 ·
allan-fraser avatar image allan-fraser snoobler commented ·

Quick update.

19:30 shore power on, MultiPlus delivering ~50A, Philippi SoC 89%.

22:15, MultiPlus delivering 10A @ 14.4v, SoC 94%.

07:30 the next day, MultiPlus delivering 1.5A @ 13.5v, SoC 100%.

The Philippi adjusts for drift by automatically setting to SoC to 100% if, for more than 4 minutes, the charge voltage is constant at Float AND the charge current is below 2% of capacity (in my case 10A).

I’ll try out the 6 hour fixed absorption option next week.

0 Likes 0 ·
allan-fraser avatar image allan-fraser commented ·
PS - I did check that with Expert Mode enabled all of the hidden parameters are the same on all three controllers.
0 Likes 0 ·

Related Resources

Additional resources still need to be added for this topic

Additional resources still need to be added for this topic

VE.Smart Networking Manual